Chocolate and depression

Tuesday April 27 2010

“The blues make you crave chocolate,” according to the Daily Mail, while the BBC reports that “Chocolate lovers are more depressive”.

The news is based on research comparing symptoms of depression with chocolate consumption levels in 931 men and women. It found that participants with high depression scores ate about 12 servings of chocolate per month. Those with low scores ate an average of 8.4 servings, and non-depressed participants ate only 5.4 servings. None were taking anti-depressants.

Both news sources emphasise that the results show a potential link between chocolate and depression. But they highlight that, by design, it was unable to say whether chocolate caused depression or the other way around. Only a large study that follows the eating habits of many people over time could test which of these theories is true.This should perhaps be the next step in chocolate research.

Where did the story come from?

This research was carried out by Dr Natalie Rose and colleagues from the University of California in San Diego. The study was funded by grants from the National Heart, Lung and Blood Institute of the US National Institutes of Health. The study was published in the peer-reviewed medical journal Archives of Internal Medicine.

The Times and Metro appropriately highlighted the finding that consuming other antioxidant-rich substances, such as fish, coffee, fruits and vegetables, had no bearing on mood. This suggests that the findings are specific to chocolate.

What kind of research was this?

This was a cross-sectional study looking at the relationship between the average quantity of chocolate eaten per week (assessed by questionnaire) and depressed mood, which was assessed by using a validated pyschological scale called the Center for Epidemiologic Studies Depression Scale (CES-D).

The depression screening scale divided participants into three groups: those with probable major depression, those screening positive for depression but not major depression, and those who were unlikely to have depression. In addition to the depression screening questionnaires, participants were asked two questions about their chocolate consumption: ‘how many times a week do you consume any chocolate?’ and ‘how many servings a month do you consume?’.

A serving was considered to be one small bar or one ounce (28g) of chocolate. Smaller and larger quantities were defined in relation to this medium serving: a small serving was half the size of a medium one, while a large serving was equivalent to one and a half times the medium.

The survey was cross-sectional and used subjective measures of chocolate consumption (estimated through questionnaires). This means that it has several limitations that make it unable to prove that chocolate causes depression or that depressed people eat chocolate to make themselves feel better.

What did the research involve?

The authors of this study say that chocolate is constantly proclaimed to have benefits on mood, but they are surprised by the lack of robust studies directly examining the link between chocolate consumption and mood in humans. To research this relationship, the authors drew data from a study that examined the non-cardiac effects of reducing cholesterol levels.

They recruited a total of 1,018 participants aged 20 to 85 years (694 men and 324 women) from San Diego. They excluded people with known vascular disease, diabetes, high/low levels of cholesterol, or those taking anti-depressants (78 people).

The participants were asked to complete food questionnaires and a depression screening questionnaire. After excluding people who did not complete both questionnaires, 931 people were available for analysis.

One food questionnaire, the SSQ-C, simply asked participants how many times a week they consumed chocolate. The second was a more intensive Food Frequency Questionnaire (FFQ-C), which asked about the absolute frequency of any chocolate consumption (times per month) and the amount of chocolate consumed (servings per month). Responses on daily or monthly consumption were converted into per-month consumption estimates to provide a measure that could be compared across the questionnaires. The FFQ also asked about other foods and nutrients, including intake of carbohydrates, fat and energy.

The researchers also administered the Center for Epidemiologic Studies Depression Scale (CES-D) tests, which asks participants about 20 symptoms of depression, and scores each of their answers on a scale of four (zero to three), giving a maximum score of 60. The scale measures depressive feelings experienced during the previous week.

The researchers analysed the data appropriately, using cut-off points to indicate minor depressive symptoms (above 16 but less than 22) and more major symptoms to indicate a depressive disorder (more than 22). Anyone scoring less than 16 was considered to be free of depression. The results of this analysis were not adjusted for the influence of other food intake, although the researchers did do similar analyses for fat, energy and carbohydrate.

What were the basic results?

The average age of participants was 57.6 years, and their average BMI was 27.8.

The average CES-D score was 7.7, ranging from 0 to 45 (maximum possible score 60). Average chocolate consumption for the whole group was six servings per month, with participants eating chocolate on six occasions per month.

Participants with a CES-D score of 16 or higher reported significantly more chocolate consumption (8.4 servings per month) compared to those with lower CES-D scores of less than 16 points (5.4 servings per month). The group with the highest CES-D scores (22 or higher) had even higher chocolate consumption (11.8 servings per month). These differences between the groups were statistically significant.

In contrast to the findings on chocolate, differences in fat, energy or carbohydrates intakes in each CES-D group were not significant. This suggests that it is specifically chocolate that has a relationship with mood rather than other foodstuffs.

How did the researchers interpret the results?

The researchers say that “higher CES-D depression scores were associated with greater chocolate consumption. Whether there is a causal connection, and if so in which direction, is a matter for future prospective study”.


This study will be of interest to many, but unfortunately it has not resolved the debate about whether depression causes people to eat chocolate or if people take chocolate to relieve low mood. The amount of chocolate eaten (six servings a month on average) may be seen as relatively little by some regular chocolate consumers. The authors acknowledge several limitations:

  • As the study was conducted for a different initial purpose, (looking at vascular disease) it is possible that some groups of participants were excluded because of vascular disease or age. This may have skewed the selection of participants, making them unrepresentative of a general population.
  • The study was based on a self-report of diet and chocolate and other nutrient consumption. This could have introduced some error or bias in that many people cannot accurately recall or estimate an average consumption of these items. As a general food frequency questionnaire was used, the participants may not have been aware of the importance of the chocolate question.
  • The CES-D screening scale is a tool for picking up symptoms that need further evaluation; it does not indicate a diagnosis of depression according to the accepted criteria. It does indicate an increased risk, however it is not correct to say that a link with ‘depression’ has been proven.
  • Different chocolate preparations were not assessed. Neither were some of the contents of chocolate that are thought to underlie the effect. The researchers mention that certain specific substances that naturally occur in chocolate (phenylethylamine, anandamine or theobromine) could be examined in future studies.

Overall, this study shows that people who screen positive on a depression screening scale eat more chocolate than those who do not. To determine whether the link is causative, people will need to be tested in long-term studies that objectively assess chocolate consumption at the start of the study and follow people to observe how depressive symptoms develop over time.

Analysis by Bazian
Edited by NHS Choices